Elasticsearch - RocketRide Documentation
Skip to main content

Elasticsearch

View as Markdown

What it does

Elasticsearch node with two modes: Index (BM25 full-text search) and Vector Store (semantic/kNN search). Switch between modes with the Store Mode toggle, no pipeline rewiring needed.

Lanes:

Lane inLane outDescription
text-Ingest raw text (index mode only)
documents-Ingest pre-embedded documents (vector store mode only)
questionstextSearch and return matching text
questionsdocumentsSearch and return matching documents
questionsanswersSearch and return matching documents as answers
questionsquestionsEnrich question with matching documents for downstream nodes

Documents must be run through an embedding node before reaching this node (vector store mode only).

Profiles

ProfileDefault portDescription
Self-managed (default)9200Your own Elasticsearch instance
Elastic Cloud Hosted9243Hosted deployment: requires host and API key
Elastic Cloud Serverless443Serverless: requires host and API key

Configuration

FieldRequiredDescription
HostyesElasticsearch server address
PortyesElasticsearch server port
API Keycloud onlyAuthentication token
Index NameyesElasticsearch index (lowercase, max 255 chars)
Store ModeyesIndex (BM25 text search) or Vector Store (semantic)

Index mode: search options

FieldDefaultDescription
Match Operatororor: any term, and: all terms, exact: phrase match
Slop0Words allowed between terms in exact phrase match
Return contextual snippetsoffHighlight matching passages in results
Snippet size250Max characters per highlight snippet

Upstream docs